Download The value of single-shot turbo spin-echo diffusion

Neuroradiology
DOI 10.1007/s00234-007-0268-3
HEAD AND NECK RADIOLOGY
The value of single-shot turbo spin-echo diffusion-weighted
MR imaging in the detection of middle ear cholesteatoma
Bert De Foer & Jean-Philippe Vercruysse &
Anja Bernaerts & Joachim Maes & Filip Deckers &
Johan Michiels & Thomas Somers & Marc Pouillon &
Erwin Offeciers & Jan W. Casselman
Received: 3 May 2007 / Accepted: 4 June 2007
# Springer-Verlag 2007
Abstract
Introduction Single-shot (SS) turbo spin-echo (TSE) diffusion-weighted (DW) magnetic resonance imaging (MRI) is
a non echo-planar imaging (EPI) technique recently
reported for the evaluation of middle ear cholesteatoma.
We prospectively evaluated a SS TSE DW sequence in
detecting congenital or acquired middle ear cholesteatoma
and evaluated the size of middle ear cholesteatoma
detectable with this sequence. The aim of this study was
not to differentiate between inflammatory tissue and
cholesteatoma using SS TSE DW imaging.
Methods A group of 21 patients strongly suspected
clinically and/or otoscopically of having a middle ear
cholesteatoma without any history of prior surgery were
evaluated with late post-gadolinium MRI including this SS
TSE DW sequence.
Results A total of 21 middle ear cholesteatomas (5
congenital and 16 acquired) were found at surgery with a
size varying between 2 and 19 mm. Hyperintense signal on
SS TSE DW imaging compatible with cholesteatoma was
B. De Foer (*) : A. Bernaerts : J. Maes : F. Deckers :
M. Pouillon : J. W. Casselman
Department of Radiology, A.Z. Sint-Augustinus Hospital,
Antwerp, Belgium
e-mail: [email protected]
J.-P. Vercruysse : T. Somers : E. Offeciers
University Department of ENT, A.Z. Sint-Augustinus Hospital,
Antwerp, Belgium
J. W. Casselman
Department of Radiology, A.Z. Sint-Jan AV,
Bruges, Belgium
J. Michiels
Siemens Medical Solutions,
Anderlecht, Belgium
found in 19 patients. One patient showed no hyperintensity
due to autoevacuation of the cholesteatoma sac into the
external auditory canal. Another patient showed no hyperintensity because of motion artifacts.
Conclusion This study shows the high sensitivity of this SS
TSE DW sequence in detecting small middle ear cholesteatomas, with a size limit as small as 2 mm.
Keywords Cholesteatoma . Diffusion . DWI . MR
Introduction
In the detection and description of the extension of a
suspected middle ear cholesteatoma, CT is still considered
the imaging method of choice. It can demonstrate ossicular
erosion and possible complications such as tegmental
disruption and fistulization through the lateral semicircular
canal. In the past few years, magnetic resonance (MR)
imaging, including diffusion-weighted (DW) echo-planar
imaging (EPI), has gained increasing importance in the
detection of acquired middle ear cholesteatoma [1, 2].
DW EPI has been shown to be accurate in differentiating
inflammatory tissue from cholesteatoma in the non-surgically
treated middle ear, as cholesteatoma demonstrates a clear
hyperintensity on DW EPI sequences in contrast to inflammatory tissue [1].
Several reports have discussed the value of DW EPI [2]
and late post-gadolinium T1-weighted sequences [3, 4] in
the detection of pre-second-look residual cholesteatoma and
postoperative recurrent cholesteatoma [5–7]. DW EPI,
however, fails to demonstrate middle ear cholesteatoma
with a size smaller than 5 mm due to susceptibility artifacts,
lower imaging matrix and relatively thick slices [2, 7].
Recent reports have highlighted the use and value of a non-
Neuroradiology
EPI-based DW sequence in the diagnosis of primary middle
ear cholesteatoma [8] and postoperative recurrent cholesteatoma [9]. The purpose of this study was to evaluate the
sensitivity of a single-shot (SS) turbo spin-echo (TSE) DW
sequence in detecting middle ear cholesteatoma and more
specifically to evaluate the size of middle ear cholesteatoma
detectable with a SS TSE DW MR imaging sequence
compared to the 5 mm size limit of the commonly used DW
EPI.
Materials and methods
Patients
Between 2 August 2005 and 12 December 2006, we
evaluated 21 patients strongly suspected clinically and/or
otoscopically of having a middle ear cholesteatoma, prior to
their planned surgery. We used the combination of late postgadolinium coronal and axial T1-weighted sequences and a
SS TSE DW sequence. In five patients, a DW EPI sequence
was also performed. Prior to MR imaging examination, all
but one patient had a CT scan as part of their preoperative
work-up. All patients were operated upon within 4 months
of MR imaging. The patients consisted of 4 women and 17
men with an average age of 36 years. Cholesteatoma
surgery was performed by one of two experienced surgeons
(T.S. or E.O.). The surgical findings, including the exact
location of the cholesteatoma, were obtained from surgical
reports and discussion with the surgeon.
Imaging technique
CT was performed on a 16-row multislice scanner (Lightspeed, GE, Milwaukee, Wis.) using an axial volume scan
(140 kV, 250 mA, 1-s rotation, 5.62 pitch, high resolution
bone algorithm) with coronal reformations. Axial slices
were acquired at a thickness of 0.625 mm, centered in a
9.6-cm field of view on the right and left ear, with a
reconstruction interval of 0.2 mm. MR imaging was
performed on a 1.5-T superconductive unit (Magnetom
Avanto, Siemens Medical Solutions, Erlangen, Germany)
using the standard head matrix coil and two 7-cm surface
ring coils. Axial 2-mm thick spin-echo (SE) T1-weighted
images were obtained with the following parameters: TR
400 ms, TE 17 ms, matrix 192×256, field of view 150×
200 mm, two averages, acquisition time 3 min 50 s.
Coronal 2-mm thick SE T1-weighted images were acquired
with the same parameters except for the matrix, which was
set at 144×256. Coronal 2-mm thick TSE T2-weighted
images (TR 3,500 ms, TE 92 ms, matrix 192×256, field of
view 150×200 mm, turbofactor 13, 12 sections, two
averages, acquisition time 2 min 41 s) and axial 0.4 mm
thick 3-D TSE T2-weighted images (TR 1,500 ms, TE
303 ms, matrix 228×448, field of view 107×210 mm,
turbofactor 37, 48 sections, one average, acquisition time
6 min 19 s) were also obtained. In all patients, 2 mm thick
SS TSE DW images were acquired in the coronal plane (TR
2,000 ms, TE 115 ms, matrix 134×192, field of view 220×
220 mm, b 0 and 1,000 mm2/s ten signals acquired, 12
sections, imaging time 4 min 2 s). In five patients, 3 mm
thick SE planar DW images were also acquired (TR 3,000,
TE 82, matrix 128×128, field of view 210×210 mm, b 0 and
1,000 mm2/s, 20 sections, six signals acquired, imaging time
2 min 14 s). Parallel imaging techniques were not used. All
images were acquired 45 min after intravenous injection of
0.1 mmol per kg of body weight gadoterate meglumine
(Dotarem, Guerbet, Roissy, France) or gadopentetate dimeglumine (Magnevist, Schering, Berlin, Germany).
Image interpretation
All images were prospectively interpreted by two radiologists experienced in head and neck radiology (J.W.C. and
A.B. with 15 and 3 years of experience, respectively, in MR
imaging of the middle ear). Both radiologists were blinded
to the patients’ identity, clinical data and CT data. Available
DW EPI were scored first. The degree of distortion was
noted. Then, all SS TSE DW images were scored by both
readers. The findings on SS TSE DW images were
correlated with standard T1- and T2-weighted MR images,
in order to determine the degree of correlation and to
exactly localize the lesion. We were able to calculate ADC
maps in 12 cases. On DW EPI as well as on SS TSE DW
images, cholesteatoma was diagnosed if a marked hyperintensity in comparison with brain tissue (b 1,000) was
found. Standard MR imaging sequences were evaluated
looking for a moderately hyperintense lesion on T2weighted images, the characteristic peripheral enhancing
cholesteatoma matrix and the central nonenhancing cholesteatoma on T1-weighted images. The size of the lesion was
measured at its maximum transverse diameter on both axial
and coronal delayed contrast-enhanced T1-weighted SE
images. All cases were classified as positive or negative,
according to the signal characteristics described above.
Welch’s test (F-test 0.045) was used to determine the
significance of differences in the ADC values between
cholesteatoma and gray matter of adjacent temporal bone.
Results
At surgery, 21 middle ear cholesteatomas were found, of
which 5 were considered congenital and 16 acquired
(Table 1). The size and location of the cholesteatomas as
determined by MR imaging were in agreement with those
F
M
M
F
M
M
M
M
M
M
M
M
F
M
F
M
M
M
M
M
M
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Left
Left
Left
Left
Left
Left
Left
Right
Left
Left
Left
Left
Right
Right
Left
Left
Right
Left
Right
Left
Right
Side
Congenital
Congenital
Acquired
Acquired
Acquired
Acquired
Congenital
Acquired
Acquired
Acquired
Acquired
Acquired
Acquired
Congenital
Acquired
Acquired
Acquired
Acquired
Acquired
Acquired
Congenital
Congenital/acquired
2
8
2
15
8
5
13
Autoevacuated
10
4
3
10
9
5
12
5
12
19
6
3
3
Size (mm)
NA
NA
−
NA
NA
−
+ (distortion)
NA
+
−
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
−
NA
NA
−
+ (distortion)
NA
+
−
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
+
+
+
+
+
+
+
−
+
+
+
+
+
+
+
+
+
+
+
+
−
+
+
+
+
+
+
+
−
+
+
+
+
+
+
+
+
+
+
+
+
−
Reader 2
Reader 1
Reader 1
Reader 2
SS TSE DW imaging
EPI DW imaging
a
Delayed T1-W imaging: T1-weighted MR imaging sequence performed 45 min after intravenous administration of gadolinium.
NA, Non applicable
Gender
Patient no.
Table 1 Identification of cholesteatomas on MR imaging
Reader 2
+ (only T2)
+
+
+
+
+
+
−
+
+
+
+
+
+
+
+
+
+
+
+
+
Reader 1
−
+
+
+
+
+
+
−
+
−
+
+
+
+
+
+
+
+
+
+
−
T2-W and delayed T1-W
imaginga
NA
NA
NA
NA
NA
NA
74
NA
72
82
67
80
74
60
88
86
59
77
98
NA
NA
Cholesteatoma
NA
NA
NA
NA
NA
NA
79
NA
86
96
82
79
80
80
72
75
79
82
94
NA
NA
Brain
ADC (mm2/s) × 10−6
Neuroradiology
Neuroradiology
found at surgery. The lesion size varied between 2 and
19 mm. There were two false-negative cases on the SS TSE
DW images. One of the false-negative cases (considered
negative by both readers on SS TSE DW as well as on T2weighted and late post-gadolinium T1-weighted MR
images) was a cholesteatoma sac that had autoevacuated
into the external auditory canal (patient 8) (Fig. 1).
The other false-negative case considered to be negative
on SS TSE DW images by both readers was a 3-mm
epitympanic congenital cholesteatoma in a young child
(patient 21). This examination was degraded by motion
artifacts in all sequences, with no clear hyperintensity seen
on SS TSE DW images. However, the cholesteatoma in this
patient was diagnosed by one of the readers on standard late
post-gadolinium T1-weighted sequences as well as on T2weighted sequences (Fig. 2). All 19 other patients showed a
nodular hyperintensity on SS TSE DW images, diagnosed
as such by both readers. The size of the smallest
cholesteatoma was 2 mm (Fig. 3; patients 1 and 3).
Disagreement was seen between the interpretations of the
two radiologists on the standard T2-weighted and late postgadolinium images in two of these 19 patients (patients 1
and 10). In these patients both readers interpreted the SS
TSE DW images as positive, but reader 1 failed to
recognize the cholesteatoma on standard T2-weighted and
late-post gadolinium T1-weighted MR images. No susceptibility artifacts were present on this patient’s SS TSE DW
images. In five patients, DW EPI was also performed. In
two of these findings of cholesteatoma, measuring 10 and
13 mm, respectively (patients 7 and 9), were interpreted as
positive on DW EPI images by both readers (Fig. 4). All
other DW EPI images were interpreted as negative by both
readers, but in these particular three patients they were
interpreted as positive on SS TSE DW images (patients 3, 6
and 10). ADC maps were calculated for 12 patients
(Table 1). In these 12 patients, there was no statistically
significant difference in the ADC value of the cholesteatoma and the gray matter of the adjacent temporal lobe (P>
0.05). For all other patients, ADC maps could not be
calculated because of technical limitations or the small size
of the lesion.
Fig. 1 Patient 8: cholesteatoma autoevacuated into the external
auditory canal on the left side. a Axial CT image on the left side at
the level of the epitympanon showing an eroded lateral epitympanic
wall (arrowheads). Note that the malleus and incus look eroded and
“thinned” in their laterolateral diameter. Compare with the normal
situation in b. The cholesteatoma has eroded the lateral epitympanic
wall and the ossicular chain. The associated soft tissue lesion typically
seen in such cases is not visible due to its evacuation into the external
auditory canal. b Axial CT image on the right side (same level as a)
showing the normal distance of the ossicular chain to the lateral
epitympanic wall (arrowheads). Note the normal aspect of the
ossicular chain. Compare with the abnormal aspect of the lateral
epitympanic wall and ossicular chain in a. c Coronal CT reformation
at the level of the cochlea on the left side showing the empty
cholesteatoma sac (white arrowheads) causing an eroded aspect of the
head of the malleus and an amputated scutum (black arrowhead).
Again, there is no soft tissue visible. Compare with the normal right
side in d. d Coronal CT reformation at the level of the cochlea on the
right side showing a normal epitympanic space (white arrowheads),
normal ossicular chain and a normal scutum (black arrowhead). e
Axial late T1-weighted MR image after gadolinium administration at
the level of the epitympanon (same level as a) showing the enhancing
inflammation and cholesteatoma matrix (arrows) around the evacuated
air-filled cholesteatoma sac. f Coronal T2-weighted MR image at the
level of the cochlea (same level as c) showing the hypointense airfilled empty cholesteatoma sac (arrows). g Coronal SS TSE DW
image (same level as c and f) showing no clear hyperintensity
Discussion
CT still remains the first imaging tool for diagnosis,
description of extension and possible complications of
middle ear cholesteatoma. However, in the presence of
associated complications, MR imaging clearly has an
additional role [10]. Several reports have stressed the role
of DW EPI in the diagnosis of middle ear cholesteatoma [1,
2], evaluation of pre-second-look residual cholesteatoma
[2] and recurrent or relapsing cholesteatoma [5–7]. DW EPI
has been shown to be accurate in differentiating inflamma-
Neuroradiology
Fig. 2 Patient 21: small (3 mm) congenital epitympanic cholesteatoma in a young child. Examination degraded by motion artifacts. a
Axial CT image at the level of the head of the malleus and incus body
showing a small nodular soft tissue lesion (arrow) protruding into the
anterior epitympanic recessus. b Coronal reformation at the level of
the anterior epitympanic recessus showing a small nodular soft tissue
lesion (arrow) eroding the malleus head. c Coronal late postgadolinium T1-weighted MR image degraded by motion artifacts at
the level of the anterior epitympanic recessus (same level as b)
showing a hypointense lesion (arrowheads) surrounded by enhancing
inflammation: small nonenhancing cholesteatoma surrounded by
enhancing inflammation. d Coronal T2-weighted image degraded by
motion artifacts at the level of the anterior epitympanic recessus (same
level as b and c) showing a nodular moderate hyperintense lesion
compatible with the small congenital cholesteatoma. e Coronal SS
TSE DW image at the level of the anterior epitympanic recessus (same
level as b, c and d) showing a rather isointense nodular lesion. No
clear hyperintensity is noted
tory tissue from cholesteatoma [1, 6]. No false-positive
findings have been reported in the literature, so hyperintensity on DW images can be considered diagnostic for
cholesteatoma [2]. It is clear that DW EPI has a role in the
visualization of the usually quite large recurrent or
relapsing cholesteatoma [5–7]. Its value in the evaluation
of residual cholesteatoma in pre-second-look ears, however,
is limited due to the often very small size of these residual
cholesteatoma pearls [2]. The size limit for visualization of
an acquired middle ear cholesteatoma using DW EPI has in
recent studies been set at 5 mm [2]. It is clear that smaller
cholesteatoma lesions can easily be missed on DW EPI
images due to the low resolution and higher slice thickness
of the sequence and its sensitivity to susceptibility artifacts.
This is probably the main reason why DW EPI seems to be
useless for the evaluation of the usually small pre-secondlook residual cholesteatoma [2]. DW EPI, however, seems
to be reliable in the evaluation of the larger recurrent
cholesteatoma [5–7]. Other reports have discussed the use
of late post-gadolinium T1-weighted MR imaging sequence
Fig. 3 Patient 3: small (2 mm) cholesteatoma, surrounded by
inflammation. CT scanning was refused by the mother of the child.
a Coronal SS TSE DW image demonstrating a very small but clearly
hyperintense nodular lesion (arrow) in the signal void of the left
middle ear. b Axial late post-gadolinium T1-weighted MR image with
a centrally located nodular very small nonenhancing lesion (arrows)
surrounded by enhancing tissue compatible with a very small
cholesteatoma surrounded by inflammation. c Coronal T2-weighted
image (same level as a) showing the very small nodular hyperintensity
(arrow) probably situated at the mesotympanon
Neuroradiology
Fig. 4 Patient 7: congenital cholesteatoma in the left petrous bone. a
Axial CT image showing a large soft-tissue lesion in the inferoposterior aspect of the temporal bone on the left side. The lesion
(arrowheads) is in close relationship to the horizontal segment of the
carotid canal. b Axial late post-gadolinium T1-weighted MR image
(same level as a) showing the hypointense nonenhancing lesion
posterior to the horizontal segment of the internal carotid artery
(arrowheads). c Coronal T2-weighted MR image at the level of the
internal auditory canal showing the moderately hyperintense congen-
ital cholesteatoma under the internal auditory canal (arrowheads). d
Coronal DW EPI image at the same level as c. Note the symmetrical
artifacts at the interface between the temporal lobe and the temporal
bone. The congenital cholesteatoma is seen as a rather small
somewhat distorted nodular hyperintense lesion medial to the middle
ear signal void (arrowhead). e Coronal SS TSE DW image at the same
level as c and d. Note the complete lack of artifacts in this image. The
cholesteatoma is seen as an oval hyperintense lesion medial to the
signal void of the middle ear (arrowheads)
in the diagnosis of pre-second-look residual cholesteatoma
[3, 4]. In the study of Williams et al. [3] the smallest
cholesteatoma detected measured 2.5 mm.
Recently, two reports have discussed the use of a nonEPI-based DW sequence instead of DW EPI in the
evaluation of middle ear cholesteatoma [8, 9]. Both
sequences appear to be DW fast SE sequences having
clearly fewer susceptibility artifacts and a higher resolution
than DW EPI. Remarkably, in the study by Dubrulle et al.,
the size limit for detecting a recurrent cholesteatoma was
set a 5 mm (equal to the size limit with the DW EPI
sequence) despite the absence of artifacts and the higher
resolution of the sequence [9].
In order to assess the sensitivity of this SS TSE DW MR
imaging sequence in the diagnosis of cholesteatoma, we
prospectively evaluated this sequence in 21 patients
clinically or otoscopically strongly suspected of having a
middle ear cholesteatoma, without a history of prior
surgery. The aim was not to differentiate between inflammatory tissue and cholesteatoma using this SS TSE DW
MR imaging sequence, as this already has been demonstrated [2], but to see what size of cholesteatoma could be
detected using this sequence. For DW imaging, we used the
coronal plane as it seems to result in fewer artifacts than the
axial plane. This sequence uses a 180° radiofrequency
refocusing pulse for each measured echo, hence reducing
the susceptibility artifacts. It allows a higher imaging matrix
and thinner sections (2 mm). Moreover, we combined the
SS TSE DW sequence with late post-gadolinium T1weighted sequences in the coronal and axial planes in
order to see if the cholesteatoma lesions were visible on SS
TSE DW images as well as on standard MR images.
Furthermore, standard MR imaging sequences were used to
locate the lesion in the middle ear.
Out of the 21 surgically confirmed cholesteatomas, 19
were diagnosed on SS TSE DW images. One cholesteatoma
was missed by both readers (false-negative) due to the fact
that the accumulated keratin (responsible for the hyperintensity on DW images) in the cholesteatoma sac had
evacuated into the external auditory canal (patient 8;
Fig. 1). The possibility of missing small retraction pockets
or evacuated retraction pockets on DW images has been
reported previously [1, 2]. Apart from this autoevacuation,
a cholesteatoma can also evacuate as a result of cleaning the
affected ear via the external auditory canal by the ENT
surgeon. Hence, MR imaging should be performed prior to
Neuroradiology
cleaning the affected ear in order to avoid the possibility
that the retained keratin has evacuated. Theoretically, the
residual epithelial lining can be seen as a small enhancing
line on T1-weighted images but in our patient this feature
was difficult to assess because of the surrounding inflammation. Only on CT images, especially the coronal
reformatted images, could the autoevacuated sac be
demonstrated. It should be kept in mind that evacuation of
the keratin out of the retraction pocket can cause falsenegative findings.
The other false-negative finding on SS TSE DW images
was caused by a small 3-mm congenital epitympanic
cholesteatoma. All sequences in this examination were
degraded by motion artifacts (Fig. 2). Probably these
motion artifacts were responsible for misregistration of the
characteristic hyperintensity of this small cholesteatoma by
smearing the signal over different pixels in different slices.
This caused the occurrence of, in retrospect, a small rather
isointense nodule on the expected location of the cholesteatoma rather than showing a clear nodular hyperintensity
(Fig. 2). However, despite the motion artifacts, one reader
diagnosed this small cholesteatoma on the standard MR
imaging sequences, as well on T2-weighted and late postgadolinium T1-weighted images. Apart from this one
particular case, standard T2 and late post-gadolinium T1weighted MR imaging sequences added no information
concerning the detection of the cholesteatoma. It is clear,
however, that localization of the cholesteatoma was easier
on standard T2- and T1-weighted MR images than on SS
TSE DW images due to the fact that anatomical landmarks
of the membranous labyrinth such as the internal auditory
canal, vestibule, cochlea and semicircular canals were much
more clearly visualized on standard MR images than on SS
TSE DW images.
The lack of clear visualization of the anatomical
landmarks of the temporal bone on SS TSE DW images
can be regarded as one of the major drawbacks of the
sequence. There is still an ongoing debate in the literature
as to whether the hyperintensity of cholesteatoma is caused
by diffusion restriction or secondary to a T2 shine-through
effect [2]. In our series, no statistically significant difference could be found between the ADC values of cholesteatoma and gray matter in the adjacent temporal lobe. A T2
shine-through effect probably explains the hyperintensity
on the SS TSE DW images. It is very striking that with this
SS TSE DW sequence even a very small cholesteatoma, up
to a size of 2 mm, could be visualized (Fig. 3). Other
“small” cholesteatomas with a size less than or equal to
5 mm (the size limit for DW EPI) were also easily detected.
It is highly probable that the lack of susceptibility artifacts,
the slice thickness of 2 mm and the higher resolution
allowed us to detect these small cholesteatomas. Even small
cholesteatomas surrounded by inflammation were seen on
SS TSE DW images as the sequence only highlights the
keratin inside the cholesteatoma (Fig. 3). Late postgadolinium T1-weighted images in patient 3 clearly
demonstrated the enhancing inflammatory tissue around a
very small nonenhancing cholesteatoma (Fig. 3). In this
patient, surgery as well as pathology nicely demonstrated
the surrounding enhancing inflammation and the centrally
located 2-mm large nonenhancing cholesteatoma. Moreover, in this particular patient, CT scanning was not
performed because of refusal by the mother of the child,
so diagnosis was confirmed based upon only the MR
imaging findings.
Although DW EPI images were also available in only
five patients, the DW EPI findings compared to the SS TSE
DWI findings confirmed previous reports that the size limit
for detection of cholesteatoma on DW EPI seems to be
5 mm [2]. Only cholesteatomas with a size greater than
5 mm were diagnosed by both readers on DW EPI images
(Fig. 4). Moreover, cholesteatomas had a distorted aspect
on the DW EPI images compared to the SS TSE DWI
images. We conclude that the SS TSE DWI sequence is
able to detect middle ear cholesteatomas as small as 2 mm.
Associating this sequence with T2-weighted and late postgadolinium T1-weighted sequences allowed us to demonstrate surrounding inflammation and to exactly localize the
lesion in the middle ear and mastoid. The fact that standard
T2-weighted and late post-gadolinium T1-weighted images
have no clear additional value in the detection of middle ear
cholesteatoma suggests the possibility of detecting suspected middle ear cholesteatoma by starting with the SS
TSE DWI sequence alone. This probably also applies to the
important subgroup of postoperative ears in which detection of a residual cholesteatoma is even more important and
more difficult. However, exact location of a residual
cholesteatoma in postoperative ears still has to be determined with the standard sequences, including late postgadolinium T1-weighted images, so both sequences appear
to be complementary. Further studies are currently running
with the aim of determining the value of the SS TSE DWI
sequence in detecting and evaluating usually very small
pre-second-look cholesteatomas and recurrent cholesteatomas after surgery.
Conflict of interest statement
of interest.
We declare that we have no conflict
References
1. Fitzek C, Mewes T, Fitzek S, Mentzel HJ, Hunsche S, Stoeter P
(2003) Diffusion-weighted MRI of cholesteatoma of the petrous
bone. J Magn Reson Imaging 15:636–641
2. Vercruysse JP, De Foer B, Pouillon M, Somers T, Casselman J,
Offeciers E (2006) The value of diffusion-weighted MR imaging
in the diagnosis of primary acquired and residual cholesteatoma; a
surgical verified study of 100 patients. Eur Radiol 16:1461–1467
Neuroradiology
3. Williams MT, Ayache D, Alberti C, Heran F, Lafitte F, ElmalechBerges M, Piekarski JD (2003) Detection of postoperative residual
cholesteatoma with delayed contrast-enhanced MR imaging:
initial findings. Eur Radiol 13:169–174
4. Ayache D, Wiliams MT, Lejeune D, Corre A (2005) Usefulness
of delayed postcontrast magnetic resonance imaging in the
detection of residual cholesteatoma in the detection of residual
cholesteatoma after canal wall-up tympanoplasty. Laryngoscope
115:607–610
5. Aikele P, Kittner T, Offergeld C, Kaftan H, Huttenbrink KB,
Laniado M (2003) Diffusion-weighted MR imaging of
cholesteatoma in pediatric and adult patients who have
undergone middle ear surgery. AJR Am J Roentgenol
181:261–265
6. Maheshwari S, Mukherji SK (2002) Diffusion-weighted imaging
for differentiating recurrent cholesteatoma from granulation tissue
after mastoidectomy: case report. AJNR Am J Neuroradiol
23:847–849
7. Stassola A, Magluilo G, Parrotto D, Luppi G, Marini M (2004)
Detection of postoperative relapsing/residual cholesteatomas with
diffusion-weighted echo-planar magnetic resonance imaging. Otol
Neurotol 25:879–884
8. De Foer B, Vercruysse JP, Pilet B, Michiels J, Vertriest R,
Pouillon M, Somers T, Casselman JW, Offeciers E (2006) Singleshot, turbo spin-echo, diffusion-weighted imaging versus spinecho-planar, diffusion-weighted imaging in the detection of
acquired middle ear cholesteatoma. AJNR Am J Neuroradiol
27:1480–1482
9. Dubrulle F, Souillard R, Chechin D, Vaneeclo FM, Desaulty A,
Vincent C (2006) Diffusion-weighted MR imaging sequence in
the detection of postoperative recurrent cholesteatoma. Radiology
238:604–610
10. Lemmerling M, De Foer B (2004) Imaging of cholesteatomatous
and non-cholesteatomatous middle ear disease. In: Lemmerling
M, Kollias SS (eds) Radiology of the petrous bone. Springer,
Berlin, pp 31–47